Don’t Let Fragmentation Add to SAN Complexity

Storage Area Networks (SANs) are a great boon to enterprises everywhere. Because a SAN removes network traffic from the production system, network traffic is freed up to accommodate day-to-day operations—themselves a heavy load. SANs generally implement multiple physical disk drives in some form of fault tolerant disk striping (RAID), and do provide a great benefit to an enterprise: because stored data does not reside directly on any of a network's servers, server power is utilized for business applications and network capacity is released to the end user.

Connecting a machine to a SAN has always been a bit of a task—it normally has to be performed manually, and with today’s heterogeneous environments there has to be considerable know-how involved in the machine’s interaction with the SAN. It becomes even more complicated, however, with the advent of virtual machines (VMs)—for each VM, a “relationship” must be established with the SAN. Since VMs can now be created and deleted on-the-fly by the users themselves, automated solutions are now appearing that will allow VMs to be automatically connected. Whether this will be a workable solution or not remains to be seen, but obviously something needs to happen to make this operation efficient.

File fragmentation already negatively affects SAN performance, if not fully addressed with an automatic solution. Physical members in a SAN environment are not read or written to directly by an application, but instead are “seen” by an application and even the OS as one single “logical” drive. When an I/O request is processed by the file system, there are a number of attributes that must be checked which cost valuable system time. If an application has to issue multiple "unnecessary" I/O requests, as in the case of fragmentation, not only is the processor kept busier than needed, but once the I/O request has been issued, the RAID hardware and software must process it and determine to which physical member the I/O request must be directed. When files are fragmented into hundreds, thousands or tens of thousands of fragments (not at all uncommon), there are obviously many more extra I/O requests. Performance slows to a crawl.

With all that must be done to keep a SAN up and running and to ensure all machines and applications are connected, IT personnel cannot afford to be chasing down and addressing symptoms of file fragmentation. Especially with the addition of VMs, there is already enough to do. Fragmentation must be constantly addressed so that is simply eliminated—a task that can only be performed with a fully automatic solution. Such a solution works invisibly, in the background, with no negative impact on system processes and—best of all—no required scheduling by IT personnel.

Don’t let fragmentation add to SAN complexity. Make sure your fragmentation solution allows you to address factors that truly need addressing.

Don’t Let Fragmentation Bring You Down from the Cloud

The last couple of years have brought the “next big platform” to the computing world: cloud computing. A true paradigm shift, cloud computing makes it possible for companies to change over from costly company-owned computing resources to performing most needed processes via simple web interfaces through facilities owned and located outside the enterprise.

The actual computing is done by vendors providing infrastructure, platforms and software as services, and is performed using server farms that spawn virtual machines on demand to meet client needs. Several heavy-hitting companies offer full cloud computing services, including Amazon, IBM, Google, Microsoft and Yahoo. As cloud computing gains broader acceptance—which is rapidly occurring—many more providers are certain to arrive on the scene.

While it would seem that a technology as lofty as cloud computing would be far beyond the simple performance problems that have plagued systems since the earliest days, it is unfortunately not true. Yes, file fragmentation is still with us—and is more of a detriment than ever.

A key component of cloud computing is the use of virtual machines. In this environment, a single drive or set of drives is supporting a number of virtual machines—and data from all of those machines is saved on the drive or set of drives. File fragmentation, which drastically slows down performance on any drive, has an even more profound effect in virtual machines.

A virtual machine has its own I/O request which is relayed to the host system. This means that multiple I/O requests are occurring for each file request—at least one request for the guest system, another for the host system. When files are split into hundreds or thousands of fragments (not at all uncommon) there are multiple I/O requests for each fragment of every file. This scenario is then multiplied by the number of virtual machines resident on any host server, then again multiplied by the number of servers. Performance is drastically slowed—and can even be stopped—for an entire computing cloud.

Such advanced technology requires advanced solutions. The only fragmentation solution that can keep the cloud aloft is one that ensures files stored at the virtual environment hardware layer are consistently and automatically in an unfragmented state. This method uses only idle resources to actually prevent a majority of fragmentation before it occurs, which means that users are never negatively affected performance-wise, and scheduling is never required. Performance and reliability virtual machines—and thus the cloud—are constantly maximized.

Don’t let fragmentation bring you down from the cloud. Ensure your cloud computing service provider is employing a fragmentation solution that will truly allow it to fly.

Fragmentation Solutions: Invisible versus “In Your Face”

Computing technology has always striven for the “totally automatic.” It certainly wasn’t always so; just look at the level of technical skill it once took to simply operate a computer. The first systems took MIT grads to simply turn them on and get answers to equations. Down through the years, they became easier to operate and required less skill, until we finally reached the PC that anyone could run.

The goal of “fully automatic” could also be said for all the various factors that go into system administration. Except for putting the physical hardware there at a desk, a new user’s desktop can now be completely set up remotely. Network loads and traffic flows can be adjusted automatically. Entire servers (virtual) can be automatically set up and run. And now, finally, the defragmentation chore can be set up to run fully automatically, and pesky file fragmentation won’t bother anyone ever again.

But wait: if you think that claim is being made about low cost or free fragmentation solutions, think again. They must be scheduled, which means use of valuable IT hours. It also means that there are many times that defragmentation is not occurring, and performance-crippling fragmentation is continuing to impact company productivity.

There are many other drawbacks to such solutions as well, especially when compared to a state-of-the-art fully automatic solution. Some require 15 to 20 percent free space in order to defragment. Many only defragments files, instead of both files and free space. In many cases, only one instance of the built-in can be run at a time.

Additionally, some have no method of reporting on defrag results or even defrag status as they operate, leaving IT personnel in the dark. Some allow no defragmentation of system and metadata files, nor exclusion of any files from defrag. They are generally “one size fits all,” addressing all types of fragmentation and sizes of drives with one defrag method.

A true fully automatic solution requires no scheduling and is always addressing fragmentation invisibly, using only otherwise-idle resources so that there is never a negative performance impact—only the positive one. The automatic solution addresses both files and free space, and only requires 1 percent free space. It tackles drives and partitions simultaneously, instead of one at a time, and also positions frequently used files for faster access. The automatic solution fully reports on defrag status and results.

Today there is even technology for preventing a majority of fragmentation before it even occurs.

The entire point of technology, going all the way back to the origin of computing, is to decrease workload. Only the fully automatic fragmentation solution accomplishes that mandate. Make sure your fragmentation chores are addressed with the invisible background technology available today, actually lowering unnecessary IT tasks and increasing IT efficiency.

The Right Fragmentation Solution Means True ROI

Return-on-investment (ROI) is a very important term to businesses. For any dollar spent, they want it proven—and then demonstrated—that there will be more than a dollar returned for that expenditure. Especially in an economic climate like that of today, an investment with an even or a minus return is extremely unpopular. An example might be a piece of equipment costing tens of thousands of dollars that seems to be requiring a thousand dollars a month to maintain, another five thousand to man and operate and that appears to be adding little to nothing of value to the final product. Another (sometimes less obvious) example is the Vice President that generates reams of inter-office memos, ties up otherwise-productive employees in lengthy meetings and contributes less than nothing to the forward motion of the company's goals.

Another very common—but often unobserved—item with no or minus ROI is the traditional method of addressing fragmentation known as scheduled defragmentation. On the surface it may sound fine: defragmentation can be scheduled to occur during off hours so that computer files can be maintained in a fragmentation-free state and computer access and performance can be maximized. But when you start examining the actual ROI of scheduled defragmentation, the proposition starts to unravel.

First, when exactly can defragmentation be scheduled so it won't interfere with users or other production? Many of today's servers operate 24X7 and cannot be interrupted or taken down for defragmentation. Second, what vital tasks are not being completed while valuable IT hours are being invested in analyzing and scheduling defragmentation?

While both the above are important points and negatively affect ROI, it's the third point that's the real kicker: is the solution actually and fully defragmenting drives? If you apply a simple bit of scrutiny, you will find the answer to be a resounding "no!" In between the scheduled runs, fragmentation is continuing to build and impact performance, and in some cases the defragmenter isn't addressing fragmentation at all. The basic ROI of a fragmentation solution is the fact that it eliminates fragmentation and increases performance all across an enterprise; if it isn't doing that, there is no ROI and it belongs on the scrap heap with that useless machine or being ushered out the back door with that superfluous Vice President.

The only solution that truly addresses today's fragmentation is fully automatic. This solution operates invisibly, in the background, utilizing otherwise-idle resources so that fragmentation is consistently eliminated. No scheduling is ever required. With such a solution comes the real ROI expected: for a relatively small investment, computer performance and reliability are actually maximized along with employee productivity. There are few investments that can be made with that kind of solid, dependable return.

The Right Defrag Solution--More Cost Effective Than a Free Solution

A utility or function that is "free" should always be examined carefully. For example, an aspiring musician may go shopping for a keyboard and find one for a decent price that has a built-in "free" drum machine. On the surface, it sounds like a great deal; they can have rhythm accompaniment as they play and it's like having your own little band. But when they get the keyboard home and actually put this "free" drum machine to work, they find that it's very limited in the sounds it will produce and in fact makes their creations sound a bit cheesy. In order to make it sound halfway good, the drum sounds will have to be run through a separate amplifier. But then they find that they can't even run the drum machine out from the keyboard separately. After countless wasted hours and endless tweaking, the musician realizes that it would have taken far less money and time—and would have produced a better result—if he or she had just invested in a professional drum machine in the first place.

Moving over into the world of computers, one finds a similar scenario: that of a free defragmenter. Again on the surface it sounds ideal. There is no up-front cost, and it will take care of that performance-crippling fragmentation problem so IT can take their attention off of it and move onto other pressing matters.

But when the time comes to actually get the defragmentation work done, a realization will dawn upon the system administrator about how “free” this defrag solution isn’t. First it has to be scheduled, many times on systems that must remain up and running 24X7. Many IT hours can be wasted in trying to fit in such schedules. Users cannot be on the system when the free defrag is running, causing more lost time and worse, income. And in between these scheduled runs, fragmentation continues to build and impact performance and reliability, which lowers the cost-effectiveness of the entire enterprise.

Then even more problems arise. IT personnel can never tell if a disk is fully defragged because there is no progress chart in the UI. Only one instance of the defragger can be run at a time. Only local drives can be defragmented.

The only true defrag solution today is one that does not require scheduling and addresses fragmentation consistently in the background. Because only idle system resources are used, there is never a negative impact on users. Such a solution means that IT can remove their attention from fragmentation at the moment this solution is installed—performance and reliability are constantly maximized. When comparing such a solution to the free defragmenter, the actual and staggering costs of the "free" utility become readily apparent.

Fragmentation and the Virtual Revolution

It hasn’t been all that long since virtual servers came on the scene, allowing entire machines to operate as software applications. This technology meant that companies hanging onto only partially used hardware servers—and sweating over the expensive space and energy they were taking up—could consolidate them as virtual servers and fully utilize machines. Virtualization has certainly meant a revolution for data centers.

But as one might expect, the revolution has not stopped with servers. Now companies are looking at how they might virtualize the scores of desktops scattered throughout an enterprise, again reducing hardware and energy consumption and simplifying management as well. The PC monitors remain with users, but become very similar to terminals connected to virtual PCs that could be located anywhere, including a company’s main data center.

It might seem that since a machine is virtual, it would not suffer from a traditional issue such as fragmentation. After all, if a machine is existing as purely data in memory, how can file fragmentation be a problem?

The answer is that the data being utilized by a virtual machine is still being saved on a hard drive. A single drive or set of drives is supporting a number of virtual machines—called “guest systems”—and data from all of those machines is saved on the drive or set of drives on what is called the “host system.” File fragmentation, which drastically slows down performance on any drive, has an even worse effect in virtual server environments. Note that fragmentation will occur no matter what is being “virtualized”—servers, PCs or even in the future, networks.

Since a virtual machine has its own I/O request which is relayed to the host system, multiple I/O requests are occurring for each file request—minimally, one request for the guest system, then another for the host system. When files are split into hundreds or thousands of fragments (not at all uncommon) it means the generation of multiple I/O requests for each fragment of every file. This action is multiplied by the number of virtual machines resident on any host server, and doing the math it can be easily seen that the result is seriously degraded performance.

Defrag is obviously crucial for virtual machines—but it must be the right defrag technology. A fully automatic defrag solution means that files stored at the hardware layer are consistently and automatically defragmented, and fragmentation is never an issue at all. Only idle resources are utilized to defragment, which means that users never experience a negative performance impact, and scheduling is never required. Virtual machine performance and reliability are constantly maximized.

Virtualization means the sky is the limit for enterprises today. Don’t let fragmentation keep you tied to the ground.

Yes, It’s Free—but Will It Do the Job?

If you ask anyone who writes advertising copy for a living, you’ll find that the word most responded to in promotional literature of any kind is the word “free.” Not surprising—who doesn’t want something useful without paying anything? The problem is that the old adage, “If it sounds too good to be true, it probably is,” usually applies.

One version of this is the “free trial.” Some trialware is full-featured and only has a time limit. In other cases, however, it is feature-hobbled and when you really need it to do the job, it won’t.

Another is simply “free software” which has been common throughout the history of the web. When you look at the functionality, however, you may find many things missing. An example would be word processing. A free word processor might be Notepad, found on all PCs. It’s fine if you’re simply writing text and don’t care about formatting, fonts or symbols, let alone spelling and grammar checking, or the ability to embed graphics and photos. For that functionality you would need to do what many do: pay for and turn to Microsoft Word.

A free virus checker would be another example. If it’s not trialware (which is probably going to have limited functionality anyway), then it most likely is not updated with all recent virus signatures and it may or may not protect your computer. For that, you’d need a professional robust application along with a subscription that always keeps it up to date and keeps your computer safe from malware attack.

The bottom line: in responding to any offer of something free, check the functionality. In most cases, you’ll find that it won’t do the job you need it to do.

In terms of functionality, a great case in point would be a free defragmenter. In looking over the features, you would most likely find that it either must be run manually or—at best—it needs to be scheduled. Most sites cannot afford to take a system down for maintenance to defragment its hard drives, as most must remain up and running pretty near constantly. The end result is that the enterprise’s computers will continue to suffer the performance-crippling effects of fragmentation, simply because defrag can be run so seldom.

The second problem would be the utility’s ability to actually defragment your drives. Does it have the needed technology to truly do the job? Many do not, especially with today’s much larger drives, enormous file sizes and sheer number of files. A defrag utility not up for the task will simply grind endlessly and never fully defragment. Once again, a company is saddled with fragmentation’s effects—despite the apparency of a “solution.”

The lesson to be learned is that “free” does not always mean “effective.” In fact, far from it. Before you decide a free utility is best, look closely at the features.

Maximum Optimization of Virtual Environments

Virtualization has revolutionized data centers in many ways. It has made it possible to create servers within minutes for specific functions—even by users. It has brought us the ability to substantially conserve on hardware resources by running multiple servers within the same hardware platform. It has made it possible to greatly reduce the physical footprints of server farms.

Interestingly, however, there is a “missing link” in the optimization of virtual resources that, if not addressed, can drastically affect virtual performance.

First, there is the issue of file fragmentation. All hard drives suffer from this malady, but virtual systems actually suffer twice as much; there is fragmentation at both the host and guest levels. Additionally, by consolidating 4 - 5 servers into one, a single storage device is forced to work overtime due to the 4 – 5 times increase in I/O traffic. The result is heavy processing bottlenecks.

Second, multiple virtual machines are sharing mutual system resources—an activity that can become a drain on performance.

Third, when virtual hard disks are set to dynamically grow, they do not then shrink when users or applications remove data. This is a costly waste of space that could otherwise be allocated to other virtual systems.

For the first problem—fragmentation—there are, of course, defragmenters that can be implemented. The advanced technology of virtualization, however, requires a more advanced solution. Technology will soon be available that actually prevents a majority fragmentation both at the guest and host levels, as fragmentation occurs in both places. This makes fragmentation a thing of the past for virtual systems, allowing their innate performance potential to be realized.

The solution to the second problem—competition for multiple resources—would be the synchronization of the complex and ongoing activity between host and multiple guest operating systems. In addition to solving file fragmentation, this would mean additional performance optimization for the entire virtual platform.

The third problem, the “bloating” of virtual hard drives, can be solved with tools that allow system personnel to monitor wasted resources and compact virtual disks when required. This facility would allow IT personnel to efficiently allocate virtual storage resources.

These problems put together equal an overall issue in virtual machine optimization. Mutually solved, maximum performance and reliability for high-traffic virtual environments can be fully exploited. When implementing virtualization—as most enterprises are today—it is wise to take these issues, and their solutions, fully into account.